^1 
48 
We have no other midge that I know of in a red clover field that 
may be mistaken for the seed-midge, tho in Europe the leaf-midge was 
originally described from red clover. Both midges affect white clover, 
however, l)iit can be distinguished by the characters given on i)age 100. 
Life History. — In central Illinois, the clover seed-midge winters 
as a full-grown larva, or as a pupa, in the soil of the clover field or in 
dead clover heads. A few warm days and a little rain bring the larvae 
to the surface of the ground a few days after the red clover has 
started on its second year's growth. Thus in 1907, in Urbana, the 
first new leaves of red clover appeared March 19 — a little earlier than 
usual — and the orange-colored larvae of the midge were fou.nd on the 
ground from March 25 to April 4, inclusive, being most numerous 
March 30 and April 2. They are by no means abundant, however, at 
this time in a clover field of the second year, and would not be present 
at all but for the precocious development of some clover heads during 
the preceding season. Some of these larvae make a cocoon and some 
do not. In either event the larva contracts in length and its integument 
hardens and becomes duller in color, forming a pupariuui, within which 
the pupa develops. One larva, found March 25, 1907, made a cocoon 
March 30, from which a female midge emerged April 12 or 13. A 
second larva, taken March 30, formed a cocoon April 2 and gave a 
female fly April 21. Both of these flies were reared indoors, it should 
be said. Out-of-doors, midges were not found until May 15, but were 
common May 23 (at which date oviposition was in progress) and at- 
tained their maximum numbers May 30. This season was a late one, 
however, owing to continuous cold weather in April, and in the years 
1903 to 1906 inclusive the dates of maximum abundance were May 24 
and 25, the time when young green clover heads also were most numer- 
ous. The larvae work in the heads during June and the first week of July, 
and leave the heads when full grown and go to the ground to pupate. 
Our earliest date for the emergence of larvae from the heads is June 20. 
Most of the larvae emerge about June 30, tho not a few wait as late as 
July 8. The pupal period at this time is three weeks or more, and the 
flies of the second generation are most abundant in the last week of 
July and the first two weeks of August, tho scattering individuals 
emerge at almost any other time between the middle of July and the 
first of September. The egg period in early August is three days, 
varying a few hours, more or less, according to temperature. Most of 
the damage to the seed crop is done during the last two weeks of Au- 
gust and the first two weeks of September, when the larvae are most 
numerous in clover heads. Most of these larvae pass the winter as 
larvae or pupae, to emerge as flies the following May. A few larvae, 
however, attain their growth early enough to produce flies in early Sep- 
tember or even later, out-of-doors. Under the sheltered conditions of 
the insectary, but without artificial heat, flies appeared as late as Octo- 
ber 10; and in a warm room flies will emerge from clover sod thruout 
the winter. It appears that the few flies of late September do not 
succeed in propagating their kind in this latitude, owing to the direct 
